No. 2.] EPITHELIUM OF DRAGONFLY NYMPHS. \\\ 



digestive secretion is given off from the cells between the nidi, 

 and never any part of it from the nidi themselves. This is as 

 shown in preceding figures. The slender tube Frenzel found 

 connecting the nidus with the digestive cavity (which made it 

 desirable for the cryptes of Basch to be rechristened driisen 

 crypt en") has no existence. What so appeared was doubtless the 

 slender prolongation of a young cell growing toward the ental 

 surface. 



(2) After witnessing the discharge of the digestive secretion, 

 there is no question as to the part' that is functional in produc- 

 ing it; and when that part is affected by a given reagent while 

 another is not, the second must be of different character. 

 Hermann's fluid blackens all those epithelial cells which are 

 discharging or ready to discharge their secretion, and leaves 

 the small cells of the nidus, the membrana propria, and muscles 

 equally pale and clear. With haematoxylin and eosin, the 

 eosin stains deeply the functional cells, including the nuclei of 

 the more elevated ones, while the haematoxylin with its usual 

 selectivity for germinative protoplasm stains the cells of the 

 nidus most deeply. 



(3) In sections favorable for showing the typical structure 

 the nuclei of the cells are arranged in rows extending from the 

 nidus obliquely upward into the folds (see Figs. 2-4); these 

 lines indicate the course of their progress toward the surface. 

 I have whole series of sections of the stage shown in Fig. 3, 

 throughout which the nuclei are beautifully festooned between 

 the elevations of the ental surface. Naturally, the nidus shows 

 as in the figure only when the section is central. An irregu- 

 larity of common occurrence is shown at the right side of 

 Fig. 4, while the left side of that figure is quite regular. In 

 nearly every case it is easy to follow two outgoing lines of 

 nuclei from each nidus. 



(4) When the accumulation of digestive secretion is consid- 

 erable, the nuclei of the nidus are numerous; after its discharge 

 one finds but few of them remaining. But if one will add to 

 the number remaining the number of the new cells which have 

 suddenly appeared beside them which are not blackened by 

 Hermann's fluid, he will have a number about equal to the 



